34
Discrete event and motion-oriented simulation for FMS R. Carelli, W. Colombo Instituto de Automatica - Universidad Nacional de San Juan (UNSJ) San Martin 1109 (o), 5400- San Juan, Argentina Tel +54-64-213303, Fax +54-64-213672 E-mail:
[email protected] R. Bernhardt, G. Schreck
Fraunhofer Institute for Production Systems and Design Technology (IPK)
Director: Prof Dr.h.c.mult. Dr.-lng. G. Spur Pascalstrafle 8-9, D - 10587 Berlin, Germany Tel +49-30-39006-0, Fax +49-30-3911037 E-mail: Gerhard Schreck@ ipk.jhg.d400.de
Abstract In this work, an off-line simulation approach for simultaneously handling the discrete event and motion orient aspects of flexible manufacturing systems (FMS) is presented. A concept for signal handling in a discrete-event simulation based on Petri Nets and a proposed interface to motion simulation is outlined. The description of motion simulation covers the introduction of a unique interface concept for all FMS components. As an example for this approach, the RRSInterface developed for robot simulation is used. Keywords Flexible manufacturing systems, discrete event simulation, Petri net, motion simulation
1
INTRODUCTION
Simulation techniques aid in optimizing flexible manufacturing system (FMS) design. A simulation program is produced in two stages (Figure 1). The first stage involves creating a formal model of the FMS. This entails abstracting physical reality as far as problem parameters allow.
L. Camarinha-Matos et al. (eds.), Balanced Automation Systems © Springer Science+Business Media Dordrecht 1995
364
Part Twelve Modeling and Design of FMS 11
The second stage converts this formal model into a simulation program with the aid of a simulation language. Using analysis of experiments with the simulation model (Figure 2), designers now seek evidence that the chosen formal model closely represents the actual system (validation) and that the simulation program satisfies the formal model (verification) [1]. Discrete simulation refers to the jumping by the simulation model time parameter from one event to the next; simulations without jumps are called continuous. Most FMS involve discrete simulation programs, although mixed forms are possible. A continuous process may activate a discrete event, for example: the end of a continuous heating process may trigger a request for a transfer trolley.
Figure 1 A formal model precedes each simulation program.
